Mutation fractions of nonsilent mutations. (A) Stacked columns show the fraction of cases with a given number of potentially relevant nonsilent variants (ie, nonsynonymous SNVs and frameshifting insertions and deletions that were linked with cancer or occurred in genes mutated in 2 or more patients) among AP/BP cases and CP cases with poor, suboptimal, or optimal responses sequenced by WES (n = 18). Differences in distribution of variants within all diagnostic groups (P = .51) and diagnostic CP subsets (P = .81) were assessed with Fisher’s exact test. (B) Fraction of cases with a given number of potentially relevant nonsilent variants among cases sequenced by panel sequencing (n = 42). P values by Fisher’s exact test: all diagnostic groups (P = .04) and diagnostic CP subsets (P ≤ .15). (C) Fraction of cases with potentially relevant nonsilent variants at diagnosis and follow-up (FU) sequenced using either WES or panel sequencing. Only cases with more than 1 sample were included (n = 40). P values by Fisher’s exact test: diagnostic CP subset (P =.003) and follow-up CP subsets (P = .02). (D) Fraction of cases with a given number of nonsilent variants among cases sequenced by WES (n = 17). P values by Fisher’s exact test: all diagnostic groups (P = .09) and diagnostic CP subsets (P = .26). (E) Fraction of cases with a given number of nonsilent variants among cases sequenced by panel sequencing (n = 40). P values by Fisher’s exact test: all diagnostic groups (P = .01) and diagnostic CP subsets (P = .08). (F) Fraction of cases with nonsilent variants at diagnosis and follow-up, sequenced using either WES or panel sequencing. Only cases with more than 1 sample were included (n = 28). P values by Fisher’s exact test: diagnostic CP subset (P = .08) and follow-up CP subsets (P = .44).